DEVELOPING A DYNAMIC DATA ANALYTICS MODEL FOR INCREASING THE EFFICIENCY OF THE PLANNING AND DECISION-MAKING PROCESSES AT KAZAKHSTAN TEMIR ZHOLY

Abstract

This study was conducted by Nazarbayev University graduate students in collaboration with National Railway Company – Kazakhstan Temir Zholy. Company generates a huge amount of data that could be processed and analyzed in order to optimize its operations. The aim of this capstone project is to develop dynamically updated data driven decision supporting tool for the company. Another aim is to develop metrics and KPIs based on the provided input data in order to adequately assess company performance. Data provided by company includes 4.6 million rows covering routes, train numbers, stations, departure dates, number of passengers travelling, number of tickets sold, day of the week information. To handle and process such amount of information data analytics approach was used. As part of research literature review was performed covering areas of capacity management, capacity utilization factors, practices to increase efficiency, metrics and KPIs and visualization approaches. Next, methodology was developed that included data gathering, data analysis, creation of dashboard in Power BI, data visualization, cost analysis of carrying one wagon. In results part seven dashboards were developed that deal with location, route, station, wagon class, utilization, excess wagons, and seasonality. Based on the findings of this dashboards the implications for practice were drawn. These include remove excess wagons, examine dashboards to take decisions on consolidating stations or modifying frequencies, investigate opportunity to shape different type of trains based on capacity needs, exploit data from wagon class to improve marketing strategy and pricing policy.